Neurological research in African Americans to reduce health disparities (K99)

Project: Comprehensive Computational Analysis of Genetic and Regulatory Differences Between Individuals With African and European Ancestries Across Four Brain Regions. Funding from National Institute on Minority Health and Health Disparities (K99 MD0169640).

MOSAIC Scientific Society: The Association of American Medical Colleges (AAMC).

In neuroscience and genomics, individuals with recent African ancestry (AA) account for less than 5% of large-scale research cohorts for brain disorders but are 20% more likely to experience a major mental health crisis. Furthermore, divergent responses to antipsychotics between AA and European ancestry (EA) have been linked to genetic differences. Understanding these genetic and/or regulatory differences between AA and EA in the human brain, is essential to the development of effective neurotherapeutics and potentially could decrease health disparities for neurological disorders.

1. Identify and characterize ancestry-related expression differences in postmortem brain from AA and EA individuals.
   Rationale: Preliminary results revealed significant transcriptional changes across the four brain regions in neurotypical controls with admixed AA associated with genetic ancestry. Hypothesis: Genetic variations influence ancestry-related transcriptional changes via alternative isoform usage. 1) Differently expressed (DE) genomic features (genes, transcripts, exons, and junctions) associated with genetic ancestry will be identified and characterized within admixed AA. 2) eQTL analysis will be preformed to identify genetic variants interacting with genetic ancestry and expression; and 3) The contribution of these genetic variants on ancestry-related expression differences will be determined.
   
2. Identify and characterize of ancestry-related epigenetic differences in the brain.
   Rationale: Transgenerational stress has been shown to affect health outcomes. Due to past policies and practices, individuals of AA may be differentially exposed to extreme stresses across generations, which are identified as potential risk factors for common disorders. Hypothesis: Epigenetic differences drive ancestry-related expression differences in the brain. 1) Differentially methylated regions (DMRs) and genetic variation associated with DNAm between ancestries will be identified and characterized using BrainSeq Consortium publicly available expression, genetic, and DNA methylation for the caudate nucleus (n=400) and dorsolateral prefrontal cortex (n=378). 2) The contribution of these DMRs and DNAm on ancestry-related expression differences will be determined.
   
3. Evaluate ancestry-related genetic and epigenetic correlations with complex traits in the post-mortem brain.
   Rationale: The integration of genomic information with complex traits has been used to improve our understanding of disease mechanism and prioritized potential therapeutic targets. Genetic differences have been linked to divergent responses to antipsychotics, suggesting genetic background is important to understanding individual disease susceptibility. Hypothesis: Epigenetic differences drive ancestry-specific complex trait associations in the brain. 1) Causal variants impacting complex trains with ancestry-related expression and DNAm will be identified and characterized. 2) Transfer learning will be applied to improve causal variant detection for AA.

Software development

dRFEtools: dynamic Recursive Feature Elimination
Technology advances have generated larger ’OMICs datasets with applications for machine learning. Even so, sample availability results in smaller sample sizes compared to features. Dynamic recursive feature elimination (RFE) provides a flexible feature elimination framework to tackle this problem. dRFEtools provides an interpretable and flexible tool to gain biological insights from ’OMICs data using machine learning.
Authors: Kynon Jade Benjamin, Apuã Paquola, and Tarun Katipalli
Pre-print DOI: https://doi.org/10.1101/2022.07.27.501227.

Collaborations

Project: Angiotensin II receptors in the human lung
Rationale: The renin-angiotensin system is one of the most well characterized integrative pathways in humans and is known to contribute to a variety of common disorders such as hypertension, chronic renal disease, and heart failure. The wide availability of agents targeting this pathway has led to an expansion of its clinical spectrum to lung disorders. Despite this widespread interest, specific localization of this receptor family in the vertebrate lung is limited. One reason is due to the general imprecision of the available antibody reagents.
Goal: Use publicly available single-cell and bulk RNA-sequencing to identify and characterize patterns of angiotensin II receptor expression in the lung at different ages.